Ikaruga is a vertical shmup published and developed by Treasure. It was first released on arcade and Dreamcast, then on a series of other platforms (GameCube, XBLA, Android, Steam). Its gameplay and scoring system uses a polarity mechanic. The present TAS uses the GameCube version as it features the most accurate emulated console. The goal is to get the highest score possible.

Game objectives

  • 2 players
  • Hard difficulty
  • Aims for maximum score
  • Emulator: Dolphin 4.0-4091

Gameplay

Ikaruga's gameplay centers primarily around the polarity mechanic. Only bullets of an opposite polarity can kill the player. Same-color bullets are absorbed and converted into energy for the game's special weapon, a homing laser. Switching the ship's polarity also changes the color of the ship's bullets, and shooting an enemy using opposite-polarity bullets will cause double damage. Killing an enemy using same-polarity bullets triggers suicide bullets toward the player.

Difficulty differences

In hard mode, killing an enemy using opposite-polarity bullets also triggers suicide bullets, but in half quantity. Moreover, some enemies fire bullets only in hard mode, and bosses have tougher bullet patterns. Overall, hard mode features more bullets, so eventually a higher score.

Scoring system

Chains

Ikaruga scoring system relies mainly on its chain system. A chain is completed when three enemies of the same color are shot in a row. The first chain gives the player 100 points, and each subsequent chain gives double points (200, 400, etc.) until the 9th chain (called max chain) which gives 25600 points. After that, each following chain gives 25600 points. If a chain is broken, the chain value starts at 100 points again. The chain status restarts at the beginning of each stage.
On a 2 players mode, each player has its own chain status. Because of this incremental chain value, it is theoretically better to give all chains to one player. For exemple, 20 chains on 1 player is 100+200+...+12800+25600*12=332700 points, but giving 10 chains to player 1 and 10 chains to player 2 yields 2*(100+200+...+12800+25600*2)=153400 points. However, in most chapters, having 2 players allows to get more chains that is possible with one player, so that both players must kill a minimum number of enemies anyway. Only in chapter 5, player 2 gets all possible chains without lowering the total number of chains.

Bullets

Players can absorb bullets of the same color, and it yields 100 points for almost every type of bullet. Exceptions are the big spheres in the beginning of chapter 4 (500 points), electrical-type bullets in the chapter 4 mid-boss and end of chapter 5 (300 points) and homing lasers of the chapter 5 boss (1000 points). Also, lasers on chapter 3 gives 100 points per frame, making a very good source of points.

Damage

Points are granted for firing enemies, and the amount is directly linked to the quantity of health removed. So whatever the way you killed the enemy, you will get the same amount of points. As (almost) every enemy is killed in the present TAS, this scoring part is rarely important. This is used during boss fights mainly, as for example the two pods on chapter 1 boss are vulnerable a few seconds before the main target, so they can be scored.

Bosses

Points are given at the end of a boss. Each player is given 10,000 times the number of seconds remaining. For chapters with mid-bosses, the number of seconds remaining of each mid-boss/boss are summed up. Because only the integer part of the second is considered, the last second of the fight can be leeched by waiting the last frame of the second before killing the boss. Except for the chapter 5 boss, it gives a higher score to kill the boss as fast as possible.

Extends

Each player receives an extend after 3,000,000 points, then every 5,000,000 points. At the end of the game, 500,000 points are given for every extend remaining for each player.

Comparison

Here is a comparison against the best known 2 players unassisted scores. All the best scores are on the normal difficulty as it is the most popular one. Being on hard difficulty instead of normal only makes a significant difference in chapter 5, during the leeching of the boss. Also, XBLA scores were not incorporated because a few differences in the level design in this version allow higher scores (in chapter 2 and 3 at least). As a side note, scores done by VTF-INO are actually double play (one person controlling the two ships)!
ChapterTAS scoreBest known unassisted scorePlayersLink
Chapter 15,398,660 (144 chains)5,116,860 (138 chains)CVM & luchttps://www.youtube.com/watch?v=i2EoYVCijO0
Chapter 28,132,820 (230 chains)7,361,540 (220 chains)VTF-INOhttps://www.youtube.com/watch?v=5Pkhuk-amL4
Chapter 310,469,400 (305 chains)9,284,120 (287 chains)VTF-INOhttps://www.youtube.com/watch?v=lbfPBI3QPw0
Chapter 410,172,780 (243 chains)9,325,360 (239 chains)CVM & luchttps://www.youtube.com/watch?v=fN8vwjej1wE
Chapter 58,879,740 (131 chains)7,427,280 (117 chains)CVM & luchttps://www.youtube.com/watch?v=eoYYFMdohYY
All50,230,20042,580,370VTF-INOhttps://www.youtube.com/watch?v=XUDd2ZdRlOE

Nach: Extremely impressive run, doing the impossible. Despite being hard to always follow the action (at least for me), the two players shielding each other was a pleasure to behold. Very nice precision showing off what TASing is all about. This run can be improved slightly by skipping the Battle Reports, but in any case, the in-game play was terrific, and the feedback was fantastic. Accepting to stars.
RGamma: Publishing...


Joined: 6/4/2009
Posts: 893
finally this game got the TAS it deserved and i hope that soon the TAS will get the star it deserve, yes vote and i agree that putting your initials at the end of the run would be more than ok, great job
Editor, Skilled player (1824)
Joined: 8/25/2013
Posts: 1200
While it's not exactly my cup of tea, it seems well optimized. Yes vote.
effort on the first draft means less effort on any draft thereafter - some loser
Post subject: Re: Yes vote, Yes for star, can I show this at a *GDQ?
Skilled player (1743)
Joined: 9/17/2009
Posts: 4986
Location: ̶C̶a̶n̶a̶d̶a̶ "Kanatah"
dwangoAC wrote:
I had watched several of the WIP encodes and seeing the full run did not disappoint. Thank you for all of the hard work. Yes vote, nomination for Star, and a request from the author - keylie, may I exhibit this TAS at an upcoming *GDQ event (now that I've set in motion the possibility of showing TAS's that are not being replayed on the original console)? Gradius didn't quite take and I want a second chance to show the broader GDQ viewership what a perfect autoscrolling shooter can look like.
But console verifying a later generation would be amazing to the viewers. :o
Post subject: Re: Yes vote, Yes for star, can I show this at a *GDQ?
Joined: 7/2/2007
Posts: 3960
jlun2 wrote:
But console verifying a later generation would be amazing to the viewers. :o
Load times for disc-based consoles are fundamentally nondeterministic, unfortunately. I suppose it's possible that some games might still be verifiable if you had some kind of feedback loop (i.e. the input mechanism includes some way of checking the current state of the game so that inputs happen at the right time after loading), but a) that assumes that nothing significant happens while loading, in particular stuff like advancing RNGs, b) it would be horrifically complicated to implement, and of course c) it'd be specific to the game you implemented it for, not a generalizable solution.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Joined: 5/24/2004
Posts: 262
Phenomenal work keylie. Would you mind sharing the GC netranking code for this run? I'm interested to see if the algorithm for encoding the score still works for a score that high. I don't think we ever located the most significant bit. :)
Samsara
She/They
Senior Judge, Site Admin, Expert player (2241)
Joined: 11/13/2006
Posts: 2822
Location: Northern California
jlun2 wrote:
But console verifying a later generation would be amazing to the viewers. :o
Does it need to be console verified to show at a *GDQ, though? Or I guess more specifically, is console verification a personal preference for the presenters or is it actually a requirement set by *GDQ officials? I'd love to see this played at a *GDQ but given how spotty console verification is even for the systems where it works well... As for the run itself? Voted yes, and echoing the Star sentiment. It's a really high quality run that shows off what a TAS can do compared to even the best of the best unassisted runs.
TASvideos Admin and acting Senior Judge 💙 Currently unable to dedicate a lot of time to the site, taking care of family. Now infrequently posting on Bluesky
warmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Post subject: Re: Yes vote, Yes for star, can I show this at a *GDQ?
Skilled player (1743)
Joined: 9/17/2009
Posts: 4986
Location: ̶C̶a̶n̶a̶d̶a̶ "Kanatah"
Derakon wrote:
jlun2 wrote:
But console verifying a later generation would be amazing to the viewers. :o
Load times for disc-based consoles are fundamentally nondeterministic, unfortunately. I suppose it's possible that some games might still be verifiable if you had some kind of feedback loop (i.e. the input mechanism includes some way of checking the current state of the game so that inputs happen at the right time after loading), but a) that assumes that nothing significant happens while loading, in particular stuff like advancing RNGs, b) it would be horrifically complicated to implement, and of course c) it'd be specific to the game you implemented it for, not a generalizable solution.
Would running it using a usb loader eliminate the nondeterminism of loading times? Also for the above, idk. :P
Fortranm
He/Him
Editor, Experienced player (879)
Joined: 10/19/2013
Posts: 1122
Easy YES vote. Amazed by keylie again.
Joined: 11/6/2011
Posts: 7
Easy Yes vote, I haven't seen such obvious Stars-bait in a while.
Banned User
Joined: 3/10/2004
Posts: 7698
Location: Finland
It makes it slightly duller when when there are no flowery bullet patterns...
Post subject: Re: Yes vote, Yes for star, can I show this at a *GDQ?
keylie
He/Him
Editor, Emulator Coder, Expert player (2844)
Joined: 3/17/2013
Posts: 392
Thanks for all the positive feedback.
dwangoAC wrote:
Yes vote, nomination for Star, and a request from the author - keylie, may I exhibit this TAS at an upcoming *GDQ event (now that I've set in motion the possibility of showing TAS's that are not being replayed on the original console)?
Yes, I would be very honered to see my TAS presented at the GQD.
Andypro wrote:
Would you mind sharing the GC netranking code for this run? I'm interested to see if the algorithm for encoding the score still works for a score that high. I don't think we ever located the most significant bit. :)
Well, it appears only 1p runs are eligible for net rankings.
Asdrien
He/Him
Player (141)
Joined: 11/17/2014
Posts: 27
Location: France
Amazing run! yes vote.
Joined: 7/2/2007
Posts: 3960
Samsara wrote:
Does it need to be console verified to show at a *GDQ, though? Or I guess more specifically, is console verification a personal preference for the presenters or is it actually a requirement set by *GDQ officials? I'd love to see this played at a *GDQ but given how spotty console verification is even for the systems where it works well...
Console verification is a great way of saying "TASes actually aren't playing a different game", and thus they make a great sort of "community outreach" project. And of course they're also helpful for verifying that our emulators are accurate. For *GDQ, though, console-verifying introduces a lot of risk to presentations, as dwangoAC well knows. It's all too easy for something to disrupt the verification process, and you don't want your audience to be watching as you try things over and over again while saying "Hm, this has never happened before."
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Moderator, Senior Ambassador, Experienced player (907)
Joined: 9/14/2008
Posts: 1014
I don't want to further sidetrack this thread from its intended purpose of discussing how awesome this run is so I'll just briefly say that yes, I plan to continue to do things at GDQ events that involve console verification / replaying TAS's on real unmodified consoles. I also plan to introduce TAS's that are not played back on real consoles to demonstrate amazing content such as this that would otherwise be either extremely risky or impossible to exhibit, for all of the reasons others have pointed out as well. Thank you, keylie, for the permission to present your game; I will let you know if and when I pitch the idea. Don't be surprised if I ask you to also be involved in some capacity. :)
I was laid off in May 2023 and became too ill to work this year and could use support via Patreon or onetime donations as work on TASBot Re: and TASBot HD is stalled. I'm dwangoAC, TASVideos Senior Ambassador and BDFL of the TASBot community; when healthy, I post TAS content on YouTube.com/dwangoAC based on livestreams from Twitch.tv/dwangoAC.
Player (13)
Joined: 6/17/2006
Posts: 510
That's a lot of points! Now I want to see a 1P version... ;)
PoochyEXE
He/Him
Active player (274)
Joined: 12/5/2010
Posts: 32
Wow. Just wow. That was absolutely brilliant! This belongs in Stars for sure, and I smell a TAS of the Year award or two as well. A yes vote isn't enough to do this justice, I need an option for my vote to be the entire "I'll have what she's having" scene from When Harry Met Sally.
Editor, Reviewer, Experienced player (980)
Joined: 4/17/2004
Posts: 3109
Location: Sweden
This was great! I vote yes and wouldn't mind this having a star. Since this is a run which aims for score, wouldn't it be a good idea to at least put the score you achieve in the submission? Perhaps per level, and a comparison with the current WR would also be interesting.
Joined: 3/3/2010
Posts: 87
I noticed that, during the battle report screens, there is a black box in the upper left corner of the window where (I assume) the battle report title goes during the tally. I also read on the Dolphin site that a similar problem had been fixed for many games on a recent revision. Can anyone who has the game verify that the run syncs on newer versions and that the issue is fixed before the inevitable publication encodes? Or is that part of the publishing process anyway for Dolphin movies?
Joined: 2/27/2011
Posts: 69
Location: Calgary, Alberta
Truncated wrote:
Since this is a run which aims for score, wouldn't it be a good idea to at least put the score you achieve in the submission? Perhaps per level, and a comparison with the current WR would also be interesting.
Yeah not knowing WR scores in comparison does take away an opportunity to show the gap between tool-assisted and unassisted to people who don't know the game. Some of the action, like Chapter 3, has the ships skirting along the top edge of the screen which does make it less interesting. Or rather, I perhaps mean that it lacks the "Oh shit!" feeling you get from playing it yourself :P Bosses too show very little mechanics, but I was surprised by the techniques used - especially for Chapter 5. I played this for GCN and was equally sad about the removal of the story. Since I knew enough of the story, when I finally played enough hours to unlock unlimited continues and beat the game, I was shocked when you sacrifice yourself and I watched the credits with my jaw on the floor. For a game like this, it had such an impact on me and I think the ending is one of the best in gaming :P I'm glad this run was finished and I did watch a couple single level encodes on YouTube during development. A really great run. The "amount of tool-assisted finesse" is definitely there for the Stars tier, but I wonder how impressive this run would be to someone who doesn't know the game at all. (And maybe that is just me doubting that it would be appropriate for the Newcomer Corner as it should be a star) And since Dot Eater runs probably would fail an entertainment value criteria for this site, I'll go search out those on YouTube :P
Editor, Experienced player (570)
Joined: 11/8/2010
Posts: 4038
PKFC wrote:
The "amount of tool-assisted finesse" is definitely there for the Stars tier, but I wonder how impressive this run would be to someone who doesn't know the game at all.
As someone who's never seen this game before, I found this really cool to watch, and a beautiful demonstration, as others have said. It was really easy to watch and enjoy, a lot like keylie's earlier Braid run. I second the nomination for Star, and vote Yes, of course!
Ambassador, Experienced player (710)
Joined: 7/17/2004
Posts: 985
Location: The FLOATING CASTLE
This is definitely S++ material.
Joined: 5/8/2010
Posts: 177
Location: Entropy
This gameplay is like watching a ballet, I love it. As I have this game I know how great this video is. Big Yes vote from me, just beautiful to see.
Former player
Joined: 6/30/2010
Posts: 1107
Location: Zurich, Switzerland
I agree with CoolKirby. I never played the game, but this was still very entertaining for me. Probably not as much as it would have been if I knew it better, but interesting enough to never get bored even once.
Current project: Gex 3 any% Paused: Gex 64 any% There are no N64 emulators. Just SM64 emulators with hacky support for all the other games.
z1mb0bw4y
She/Her
Joined: 11/26/2012
Posts: 70
Gotta chime in to ditto all the comments about having never played this one but fully enjoyed it. I'll be buying and playing this game in the near future. Yes vote and +1 for star consideration. Also very hype to potentially see this shown off at a GDQ event! Looking forward to that one
keylie
He/Him
Editor, Emulator Coder, Expert player (2844)
Joined: 3/17/2013
Posts: 392
I just added in the description a comparison with best known unassisted 2p scores. If someone has a knowledge of better ones, feel free to tell.